Download Economic-Base Theory CHAPTER 3

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Transcript
Economic-Base Theory
Chapter 3
CHAPTER 3
REGIONAL MODELS OF INCOME DETERMINATION: SIMPLE
ECONOMIC-BASE THEORY ................................................................... 1
ECONOMIC-BASE CONCEPTS ......................................................................... 1
Antecedents .............................................................................................. 1
Modern origins......................................................................................... 3
THE STRUCTURE OF MACROECONOMIC MODELS ........................................... 3
THE "STRAWMAN" EXPORT-BASE MODEL...................................................... 5
THE TYPICAL ECONOMIC-BASE MODEL ......................................................... 6
TECHNIQUES FOR CALCULATING MULTIPLIER VALUES .................................. 8
Comparison of planner's relationship and the economist's model.......... 8
The survey method ................................................................................... 8
The ad hoc assumption approach ............................................................ 8
Location quotients.................................................................................... 9
Minimum requirements .......................................................................... 11
"Differential" multipliers: a multiple-regression analysis ................... 11
CRITIQUE: ADVANTAGES, DISADVANTAGES, PRAISE, CRITICISM ................. 12
APPENDIX A REVIEW OF ECONOMIC-BASE LITERATURE ..... 13
APPENDIX B AN ECONOMIC-BASE MODEL OF ATLANTA ....... 33
NOTE A. TECHNIQUES FOR DATA ANALYSIS .............................. 37
INTRODUCTION ........................................................................................... 37
LOCATION QUOTIENTS ................................................................................ 37
SHIFT-SHARE ANALYSIS .............................................................................. 38
THOUGHTS ON WRITING AN AREA PROFILE ................................................. 39
ELEMENTS TO INCLUDE IN A LOCATION QUOTIENT ANALYSIS ..................... 40
WASchaffer
Draft 5/11/2010
Economic-Base Theory
Chapter 3
REGIONAL MODELS OF INCOME
DETERMINATION: SIMPLE ECONOMICBASE THEORY
will then look at methods of estimating
the values of multipliers.
Economic-base concepts
Economic-base concepts originated with
the need to predict the effects of new
economic activity on cities and regions.
Say a new plant is located in our city. It
directly employs a certain number of
people. In a market economy these
employees depend on others to provide
food, housing, clothing, education,
protection and other requirements of the
good life. The question which city
planners and economists need to answer,
then, is "what are the indirect effects of
this new activity on employment and
income in the community?" With these
estimates in hand, we can work toward
planning the social infrastructure needed
to support all of these people.
Economic-base models focus on the
demand side of the economy. They
ignore the supply side, or the productive
nature of investment, and are thus shortrun in approach. In their modern form,
they are in the tradition of Keynesian
macroeconomics. In an introductory
economics course, we might start with a
simple model of a closed economy,
usually with some unemployment. In
regional economics we deal with an open
economy with a highly elastic supply of
labor.
Antecedents
We commonly divide economies into
two often opposing parts. In action, it's us
against them; in primitive life, it is hunters
and gatherers; in analysis, it will be
primary and secondary, productive and
nonproductive, basic and nonbasic, export
and support, fillers and builders,
productive and sterile workers, necessary
and surplus labor, etc. The following
notes trace obvious antecedents.1
Mercantilistic thought is a prime
example. During the period in which the
mercantilists were dominant, normally
considered to be from 1500 to 1776, the
nation-states of Europe were
consolidating their power and gaining
strength to resist or conquer others. The
writers who documented the times
emphasized a philosophy not unlike that
of a modern merchant or chamber of
commerce.
The mercantilists stressed accumulating
a supply of gold with which to pursue the
nation's political and military objectives.
The economic base of a nation included
the sectors which created a favorable
balance of trade. Goods were produced
for export despite the needs of a poor
population, export of unprocessed
materials was prohibited, shipping in local
It is appropriate to start this chapter
first with a look at the place of economicbase theory in the history of economic
thought and proceed to a review the
simple Keynesian model and the
elementary economic-base models. We
WASchaffer
1 This section is based on (Oser 1963) and (Kang and
Palmer 1958). Oser's The Evolution of Economic
Thought is one of the best short histories of economic
thought in print.
1
Draft 5/11/2010
Economic-Base Theory
Chapter 3
bottoms was forced whenever possible,
and colonies were exploited as a source of
raw materials.
were sterile, simply passing value on to
consumers. This classification of
productive and sterile activities is similar
to the basic and service classification in
economic-base discussions.
Thomas Mun, a merchant in the Italian
and Near Eastern trade and a director of
the East India Company, was probably the
most famous of these writers. His
exposition of mercantilist doctrine in
England's Treasure by Foreign Trade,
written in 1630, explained how "… to
enrich the kingdom and to encrease our
Treasure." He emphasized a surplus of
exports as the key:
And second, the Physiocrats visualized
money flowing through the economic
system in much the same way as blood
flows through the living body. Quesnay's
tableau economique was a predecessor of
the circular-flow diagrams popularized in
Keynesian macroeconomics.
Adam Smith, writing in 1776, and
heavily influenced by these French
authors, took a less extreme but
nevertheless strong position. He
emphasized production of material or
tangible goods and considered service and
government as unproductive.
Although a Kingdom may be enriched by gifts
received, or by purchase taken from some other
Nations, yet these are things uncertain and of
small consideration when they happen. The
ordinary means therefore to encrease our wealth
and treasure is by Forraign Trade, wherein wee
must ever observe this rule; to sell more to
strangers yearly than wee consume of theirs in
value.(from Oser 1963 p.14)
Karl Marx, in das Kapital, also divided
the economy into two parts. To Marx,
necessary labor was the source of wealth
and was paid for with a wage barely
sufficient to maintain its provider.
Surplus labor was also provided by
workers but its value was appropriated by
the capitalists in the form of surplus value.
Workers had to produce not only what
they consumed but also a surplus for the
capitalist. Menial servants, landlords, the
Church, and commercial activities were
unproductive – they added nothing to total
value.
The Physiocrats, led by François
Quesnay and briefly prominent in France
in the second half of the 18th century
prior to the French Revolution, responded
to the excesses of the mercantilists with
several points important to later thought.
They considered society subject to the
laws of nature and opposed governmental
interference beyond protection of life,
property, and freedom of contract. They
opposed all feudal, mercantilist, and
government restrictions. "Laissez faire,
laissez passer," the theme phrase for the
free enterprise system, is from the
Physiocrats. They opposed luxury goods
as interfering with the accumulation of
capital.
Others of the nineteenth century were
more generous. Jean Baptiste Say in his
Treatise on Political Economy (1803)
popularized Adam Smith in France. Say's
famous Law of Markets, paraphrased as
"supply creates its own demand," required
that all work be productive, that all
compensated activity creates utility.
But, for our purposes, they were
precursors of economic-base thought in
two ways. First, they were important in
their treatment of the sources of value. To
the Physiocrats, only agriculture was
productive. The soil yielded all value;
manufacturing, trade, and the professions
WASchaffer
Nevertheless, we can see a strong line
of thought dividing economic activities
into two parts, and we can see economic-
2
Draft 5/11/2010
Economic-Base Theory
Chapter 3
base concepts as fitting into a centuriesold pattern.
assumptions, equilibrium conditions, and
finally the solution. Since this is a process
we will follow with each new model
considered, it may be worthwhile to
review the nature of these model
elements.
Modern origins
Modern literature on the economic base
has been voluminous, but plagued
occasionally by scholastic sloppiness in
appropriate citations.
A definition is a statement of fact. By
definition, it is always true. In
mathematics, the proper term is identity.
One of the more important identities in
macroeconomics is the national income
identity: realized national income (actual
expenditures) is the sum of realized
consumption and realized investment. In
the simple national model, this has to be a
true statement—it is a tautology. Actual
expenditures have to equal their sum!
It seems that Werner Sombart, a
German (historical) economist writing in
the early part of this century, should
receive major credit for modern concepts.2
Sombart was responsible for the
distinction between "town fillers" and
"town builders," ("Städtegründer" and
“Städtefüller") which appeared in
Frederick Nussbaum's A History of the
Economic Institutions of Modern Europe
(with full permission). But in a series of
articles in the early 1950's, Richard B.
Andrews quoted extensively from
Nussbaum without mentioning the fact
that Nussbaum had based his book on
Sombart's work. Andrew's work was
widely circulated and became the standard
reference.
Another important identity in the
simple model is that income (which is
another term for 'output') is equal to the
sum of consumption and savings. We, as
recipients of incomes, either spend our
incomes or we save (don't spend). This
identity can also be taken as a definition
of saving as the difference between
income and consumption.
Behavioral assumptions are equations
describing the behavior of certain groups,
or actors, in the economy. In this case,
the key behavioral relationship is the
consumption function, which postulates
consumption as dependent on, or caused
by, income:
The structure of macroeconomic
models
It is convenient to begin with a review
of the basic elements of model building.
We can start with the simplest of all
macroeconomic models, the Keynesian
model of a closed economy. This model
is presented algebraically in Illustration
Error! Reference source not found..1
and follows the standard format we will
use in all of our models: we outline
definitions, behavioral or technical
C = f(Y)
which in its linear form may be expressed
as:
C = a +cY
where a represents autonomous
consumption and c is the marginal
propensity to consume (dC/dY). The
parameters of the equation are a and c.
Recall that if a>0, dC/dY<C/Y.
2 I rely on Günter Krumme for this statement (Krumme
1968). On his excellent web site Professor Krumme
points out that, according to Marc de Smidt, Sombart
himself traces the concept back to a 1659 manuscript by
the Dutch mercantilst Pieter de la Court. See
http://faculty.washington.edu/~krumme/papers/sombart.html
WASchaffer
3
Draft 5/11/2010
Economic-Base Theory
Chapter 3
An incidental but important result of
this assumption is that saving is also a
function of income:
investment, I, is determined outside the
system. It is planned. In terms common
to model building, it is an exogenous
variable in contrast to consumption, which
is determined endogenously (that is,
'within the system').
S  Y - C = -a + (1 - c)Y
The other important behavioral
assumption in this simple model is that
Illustration Error! Reference source not found..1 The simple Keynesian
model
Definitions or identities:
Planned Expenditures  Consumption + Investment (planned sources of income)
(1)
E  C+I
Actual Income  Consumption + Savings (actual disposition of income)
(2)
Y  C+S
Behavioral or technical assumptions:
Consumption = A linear function of income (both planned and actual)
(3)
C = a + cY
(c < 1 = the marginal propensity to consume)
Investment = Planned investment (an exogenously determined value)
(4)
I = I'
Equilibrium condition:
Income = Expenditures, or actual income is equal planned expenditures
(5)
Y=E
or, with C + S = C + I, we can subtract C from both sides to form an
equivalent equilibrium condition:
Drains = Additions
(6)
S=I
Solution by substitution:
Y=C+I
Substitute (1) into (5)
Y = a + cY + I'
Substitute (3) and (4)
Y - cY = a + I'
Gather the Y, or income, terms
(1 - c)Y = a + I'
Factor out Y
Y = {1/[1 - c]}*(a + I')
Isolate Y through division
The simple Keynesian investment multiplier is:
dY/dI = 1/[1 - c]
(An example of a technical
assumption in economics is the
production function. A production
function describes the relations between
WASchaffer
inputs and outputs. A familiar example
is Q=F(K,L), commonly used to describe
how capital and labor are combined to
produce output.)
4
Draft 5/11/2010
Economic-Base Theory
Chapter 3
Equilibrium is a condition in which
the expectations (plans) of decisionmakers (actors) in the system are met.
In this simple model, the equilibrium
condition is that income equals planned
expenditures, or, what is the same thing,
that saving (which sets the limits on
actual investment) equals planned
investment.
literature. The "export-base" model, in
which the sole determinant of economic
growth is exports, is often built to
represent the arguments of other
practitioners. However, you can seldom
find an "export-base" theorist who is not
also an "economic-base" theorist readily
acknowledging many other determinants
of growth than exports alone.
The point is that planned investment
and saving do not have to be equal (even
though, in the end, actual saving has to
equal actual investment—this is a
fundamental principle of accounting).
When they are equal, then all parties are
satisfied. When they are not, forces are
at play which will take income to a
lower or higher level, bringing saving
into equality with planned investment.3
Now let us construct this strawman
and see how a pure export-base stance is
untenable. We move into an open
economy and make exports the sole
exogenous factor. If any autonomous
expenditure is included (the easiest is for
consumption), then regional income can
exist even when exports are zero (Ghali
1977).
Presented in Illustration 3.2, the
model differs only slightly from the
simple Keynesian model. With Keynes,
the key leakage was savings. He
explained the underemployment of a
depressed economy as resulting when
planned investment fell below fullemployment equilibrium levels due to a
lack of confidence in investment
markets. His endogenous variable was
consumption, through which most
income flows occurred—the flows
became disconnected in the savinginvestment path.
Good introductions to the art of
model-building can be found in several
readily available books (e.g. Bowers and
Baird 1971; Kogiku 1968; Neal and
Shone 1976). The simple Keynesian
model is outlined in almost all texts on
the principles of economics. A good
reference is (Case and Fair 1994).
The "strawman" export-base model
It is common in economics to
construct a "strawman" against which to
rail and argue. Nowhere is this practice
more common than in the regional
In the export-base model, the
endogenous flow remains consumption,
redefined now as “domestic
expenditures.” We completely ignore
saving and hide investment expenditures
within domestic expenditures (we are
concerned not about explaining
depression in the whole economy but
about explaining changes in regional
income). The function of saving in
creating a leakage from the economy is
now assumed by imports, which is
defined as a function of income. The
3 This paragraph brings “Say’s Law” into play. Stated
by Jean Baptiste Say in the early 1800s as the “Law of
Markets,” the idea that “supply creates its own
demand” was named in 1909 by Frederic Taylor.
Keynes succinctly restated it as above and argued that
it did not apply. In Say’s time, since saving and
investment were often done by the same landed
people, it might have been more valid. But in modern
times with complex banking systems, saving is done
by many people who do not buy capital goods and
investment is done by people who borrow those
savings. So the possibility of actual savings differing
significantly from planned investment became real.
WASchaffer
5
Draft 5/11/2010
Economic-Base Theory
Chapter 3
function of investment is now assumed
by exports, the driver of the exportbased economy.
This model obviously stresses
openness and dependence of the region
on events beyond its reach.
Illustration 3.2 The pure export-base model
Definitions or identities:
Total expenditures  Domestic expenditures + Exports (inflows)
(1)
ED+X
Income  Domestic production +Imports
(2)
Y  D + M, or D  Y - M
Behavioral or technical assumptions:
Imports = a linear function of income
(3)
M = mY
(m<1, the marginal propensity to import)
Exports = an exogenously (outside-region) determined value
(4)
X = X'
Equilibrium condition:
Income = Total expenditures
(5a)
Y=E
or
Drains = Additions
(5b)
M=X
Solution by substitution:
Y=Y-M+X
Substitute (1) and (2) into (5a)
Y = Y - mY + X'
Substitute (3) and (4)
Y - Y + mY = X'
Gather the Y, or income, terms
mY = X'
Reduce
Y = (1/m)*X'
Isolate Y through division
The export-base multiplier is:
dY/dX = 1/m
The missing element is autonomous
consumption (which appeared in the
simple Keynesian model). Whether or
not it is included seems to me to be a
matter of personal preferences. On the
one hand, it might be nice to be
complete and consistent with the
Keynesian model. In addition, it serves
to warn us that the consumption function
is probably curvilinear, originating at the
origin and rising at a decreasing rate
with respect to income. The marginal
The typical economic-base model
To make the model slightly more
realistic (or, rather, less simplistic!),
saving and exogenously determined
investment can be added back into the
system.
Illustration 3.3 includes these to
develop an almost typical economic-base
model. Only minor interpretive
comments are required.
WASchaffer
6
Draft 5/11/2010
Economic-Base Theory
Chapter 3
propensity to consume at the range of
incomes over which we might work is
less than the average propensity to
consume. A positive autonomous
consumption permits us to simulate this
case.
multiplier is identical to that which
would be calculated for autonomous
consumption—we have the results
without the bother. While this is a logic
which might reduce a model to pulp if
pursued too rigorously, I have left
autonomous consumption out of this
illustration.
On the other hand, we already have
one exogenously determined nonexport
variable, investment. The investment
Illustration 3.3 The pure economic-base model
Definitions or identities:
Total expenditures  Domestic production + Exports + Investment
(1)
ED+X+I
Income  Consumption + Saving
(2)
YC+S
Consumption  Domestic expenditures + Imports
(3)
C  D + M, or D  C - M
Behavioral or technical assumptions:
Consumption = a linear function of income
(4)
C = cY
(c = the marginal propensity to consume)
Imports = a linear function of income
(5)
M = mY
(m = the marginal propensity to import)
Exports = an exogenously (outside-region) determined value
(6)
X = X'
Investment = an exogenously (outside-system) determined value
(7)
I = I'
Equilibrium condition:
Income = Total expenditures
(8a) Y = E
or
Drains = Additions
(8b) M + S = X + I
Solution by substitution:
Y=C-M+X+I
Substitute (1) and (3) into (8a)
Y = cY - mY + X' + I'
Substitute (4), (5), (6) and (7)
Y - cY + mY = X' + I'
Gather the Y, or income, terms
(1 - c + m)Y = X' + I'
Factor out Y
Y = {1/[1 - (c - m)]}*(X' + I')
Isolate Y through division
The economic-base and investment multipliers are:
dY/dX = 1/[1 - (c - m)], and dY/dI = 1/[1 - (c - m)]
WASchaffer
7
Draft 5/11/2010
Economic-Base Theory
Chapter 3
The survey method
Techniques for calculating
multiplier values
Of course, the most straight-forward
method is simply to ask businesses in the
area to specify how much of their
revenues is basic and to use their
responses to accurately divide local
business activities into basic and service
components. In practice, this is seldom
done.
Comparison of planner's relationship
and the economist's model
Concentrating purely on the practical
need to develop an easy way to forecast
community change, early planners
developed economic-base ratios (T/B for
the average ratio, and T/for the
marginal ratio, where the letters
represent total (T) and basic (B) income
(or employment) by pure observation as
rules of thumb. By 1952, economists
(Hildebrand and Mace 1950) had
developed export-base models in the
same analytic framework as the
Keynesian macroeconomists, with
multipliers expressed as (1/(1-PCL),
where PCL represents either the average
propensity to consume locally produced
goods (APCL) or the marginal
propensity (MPCL). Could these
approaches be equivalent? Yes. Charles
M. Tiebout showed us how (Tiebout
1962). Tracing the metamorphosis for
average propensities,
The neglect of the survey approach is
easy to explain. It is the most expensive
and time-consuming of approaches.
Questionnaires on sensitive issues such
as revenues, employment, and markets
are seldom answered freely; to obtain
even a smattering of responses the study
team must resort to personal interviews.
And even then, the interviewers must be
skilled and persuasive.
In addition, if the area is of any size,
the survey would require careful
planning. A canvass would be
prohibitive and the sample must be
carefully stratified and selected to
represent the broad spectrum of
activities represented in modern
communities. Such care and expense
would meet the test of rationality only if
data collection were in the context of a
much larger study. The limit to the
value of a simple export-base ratio is
fairly low, in the hundreds of dollars.
T/B = 1/(B/T) = 1/((T-NB)/T)) = 1/(1NB/T) = 1/(1-APCL)
Here, the ratio of nonbasic activity to
total activity (NB/T) is the equivalent of
the average propensity to consume
locally produced goods.
A final argument against this simple
approach is that the survey would
probably yield data for only one year,
leading to calculation of an average
multiplier when a marginal multiplier is
the most appropriate.
So, if we can obtain values of total
and basic variables over a period of
years, we can estimate marginal exportbase multipliers by regressing the total
on the basic values. With the regression
line formulated as T = a +bB, the slope b
is the marginal multiplier (T/ for
the region.
WASchaffer
The ad hoc assumption approach
The easiest and least expensive of
methods is simply to rely on arbitrary
assignment of activities to basic or
nonbasic categories. This could be done
8
Draft 5/11/2010
Economic-Base Theory
Chapter 3
by assignment of, say, employment or
payrolls for entire industries into
categories, or it could be accomplished
with a little more finesse by estimating
proportions of employment involved in
basic activities.
Surplus or export employment in
industry i can be computed by the
formula
EXi = (1 - 1/LQi)*ei , LQi > 1,
which is easily shown to be the
difference between actual industry
employment in the area and the
"necessary" employment in the area.
Needless to say, the chance of errors
is large even for experienced analysts,
and the multiplier will again be an
average one with limited use in
analyzing the effect of change.
In fact, then, excess employment can
be computed without reference to
location quotients through this reduction
of the formula:
Location quotients
The location quotient is probably
responsible for the long life and
continuing popularity and use of
economic-base multipliers. These
quotients provide a compelling and
attractive method for estimating export
employment (or income).
EXi = ei - (Ei/E)*e
It is convenient to retain the initial
formula as a reminder of the logic, and
to compute location quotients as
reminders of the strengths of exporting
industries.
Now it is easy to estimate export
employment for each industry in the area
and to sum these estimates to yield a
value for export employment for the area
in some particular year. With this
number and total employment, an
average multiplier for the area can be
computed. With a set of these values
over 10-20 years, the more acceptable
marginal multiplier can be estimated by
simple regression.
A location quotient is defined as the
ratio
LQi = (ei/e)/(Ei/E),
where ei is area employment in industry
i, e is total employment in the area, Ei is
employment in the benchmark economy
in industry i, and E is total employment
in the benchmark economy. Normally,
the "benchmark" economy is taken to be
the nation as the closest available
approximation to a self-sufficient
economy.
While it is common to use
employment as the primary basis for
these calculations, other measures such
as wages and salaries are just as
appropriate. Indeed, wage data is more
accessible electronically, especially on
CD-ROM. County Business Patterns, a
standard source of employment and
payroll data, is available for years since
1986, two years per disk. In
considerable detail, this is the best data
for recent years, but skill with
mainframe computers, tapes, and
programming is required to gain access
Assuming that the benchmark
economy is self-sufficient, then a
location quotient greater than one means
that the area economy has more than
enough employment in industry i to
supply the region with its product. And
a quotient less than one suggests that the
area is deficient in industry i and must
import its product if the area is to
maintain normal consumption patterns.
WASchaffer
9
Draft 5/11/2010
Economic-Base Theory
Chapter 3
for earlier years. The Regional
Economic Information System (REIS),
updated on CD-ROM annually by the
U.S. Department of Commerce with a
two-year lag, includes a relatively
aggregated 16-category employment
series for the years 1969-2000 as well as
a more detailed earnings series for every
county in the nation (categories are
based on the old Standard Industrial
Classification (SIC) system). This data
makes earnings-based location quotients
a snap, especially if historic estimates
are desired. The REIS files released
June 2009, can be downloaded free of
charge from
http://www.bea.gov/regional/docs/reis20
07dvd.cfm .
of intermediate goods by producers
differ for regions depending on industry
mix. (It turns out that we can account
for industry mix with input-output
models, so this difference has been
accounted for by the march of time.).
The constant-labor-productivity
assumption is difficulty to avoid. Its
impact can be ameliorated slightly
through using earnings data, which can
be assumed to reflect regional
productivity variation through
differences in wage rates. (This
assumption could in turn be attacked if
wages vary more by area cost of living
than by productivity.)
The assumption that local demands
are met first by local production is the
more tenuous of the three. It is
obviously not true, as any visit to a
grocery or clothing store will attest. But
it is common, and a better alternative is
hard to come by.
From 2001 on, the industry categories
are based on the new North American
Industry Classification System (NAICS),
with 23 categories of employment and
even more categories for earnings. This
shift in industry definitions means that
categorical data is not available in time
series. Everything starts anew in 2001.
In addition to the disadvantages
accruing from these assumptions,
another major fault is that the method is
dependent on the degree of aggregation
of the data, making comparisons among
various studies of little value. To
illustrate the problem, consider the food
and kindred products industry in Atlanta.
The location quotient computed for this
broad industry should be less than one,
and if excess employment were
computed based on this classification,
none would be credited to the food
industry. But if the classification were
more detailed, the soft-drink industry
would show a large number of excess
employees, since the headquarters of
Coca-Cola is in the city.
Location quotients have been in use
by regional analysts for over 50 years
now, and have been commented on at
length. We should look at the
assumptions involved in their use as well
as the advantages and disadvantages.
The literature records at least three
specific assumptions: (1) that local and
benchmark consumption patterns are the
same, (2) that labor productivity is a
constant across regions, and (3) that all
local demands are met by local
production whenever possible.
The first assumption is not serious:
not only can we not discern differences
in consumption patterns without
extraordinary expense but we can
suspect that differences in production
patterns are more important. Purchases
WASchaffer
The overpowering advantages of
using location quotients are that the
method is inexpensive and the exercise
of computing excess employment may
10
Draft 5/11/2010
Economic-Base Theory
Chapter 3
give the analyst an opportunity to gain
insights of interest in themselves.
"Differential" multipliers: a multipleregression analysis
Another approach which has been
used in estimating economic-base
multipliers is to fit a multiple-regression
equation to regional data. The first of
these studies arose in a study of the
impact of military bases on Portsmouth,
New Hampshire in 1968 by Weiss and
Gooding (Weiss and Gooding 1968).
Minimum requirements
In the 1960's, when available
computing technology favored frequent
use of economic-base models, one of the
alternatives to the use of location
quotients was the minimumrequirements approach (Ullman and
Dacey 1960). This variation involved a
slight revision of the location-quotient
formula to
Simple economic-base models ignore
the possibility that different industries
may have different impacts on their
communities. The regression technique
eliminates this simplifying assumption.
Weiss and Gooding set up an equation
min
EXi = ei - (Ei/E) *e ,
min
where (Ei/E) is the minimum
employment proportion for industry i in
cities of size similar to the subject city.
You can readily see that we have
substituted a varying benchmark
employment proportion for a constant
one:
min
LQi = (ei/e)/(Ei/E)
S = Q + b1 X1 + b2 X2 + b3 X3,
where S represents service employment,
Q is a constant, and the X terms are, in
order, private export employment,
civilian employment at the Portsmouth
Naval Shipyard, and employment at
Pease Air Force Base.
.
While still appearing in various forms
in the literature, the method suffers from
two major criticisms. One is that, if
enough cities are included in the selected
set, all regions will be exporting and
none may be importing. The other is
similar in that, if we use data defined in
a fine level of detail (which should be
an improvement, as it was in locationquotient estimates), we may reduce local
needs to near zero and make almost all
production for export (Pratt 1968).
With data fitted from 1955-64, their
results were
S = -12905 + .78 X1 + .55 X2 + .35 X3
(.31)
(.14)
The multipliers are 1+ bi for each
sector.
Weiss and Gooding used a mixture of
assumption and location quotient
methods in allocating export
employment and assumed that the export
sectors were independent and that
workers in the export sectors demanded
similar services.
At any rate, the method is not
commonly used now. The locationquotient method remains the virtually
sole survivor as a simple means of
identifying export industries.
WASchaffer
(.23)
This variation on economic-base
modeling has not fallen into widespread
use for several reasons: its flexibility (in
number of exogenous sectors) and the
statistical significance of coefficients are
limited by the number of observations
available; determining the export
11
Draft 5/11/2010
Economic-Base Theory
Chapter 3
content of industry employment remains
a demanding chore; and with the rise of
desktop computing, input-output models
are better sources of industry-specific
multipliers and are similar in cost.
An employment multiplier is often used to
discuss income changes. (But this assumes
that employment and per capita income are
perfectly correlated -- in a simple economy
with perfectly elastic supplies of labor this
might be the case although, of course, the
world is not simple.)
Critique: advantages,
disadvantages, praise, criticism
Assumes that exports are the sole determinant
of economic growth. (It is not reasonable for
us to take the rap for this.) Any rational
person can see that the determinants of growth
are many -- the simple model just emphasizes
one determinant. Perhaps the fault lies in early
attempts to formulate multipliers and the ease
with which the simple multipliers could be
constructed. (Ghali 1977; Sirkin 1959))
Economic-base models suffer from
old age: they have been built by so
many analysts with varying levels of
quality and they have been criticized so
often that little remains except the
concept.
Direction of dependence may be questionable:
which comes first, export growth or a strong
service sector, or interdependence? Should
we be concerned with preconditions for export
growth (setting up an attractive service sector)
in this simple model? Are we planning growth
or explaining the true basis?
The indictment would include the
following phrases:
Short run
Nonspatial
Simple adaptations of national models
Although castigated for decades, the
economic-base model has survived as a
very succinct expression of the power of
demand in regional income
determination. The most current, and
perhaps the clearest and most complete,
statement of its status is found in a
recent review by Andrew J. Krikelas
(1992), reprinted below with permission
from the Atlanta Federal Reserve Bank..
Data is normally available for administrative
units (counties) which may be poorly defined
as economic regions.
Ignores capacity constraints
Assumes perfect elasticity of supply for inputs
Pits the area against the rest of the world,
showing no interdependence between regions
Multiplier varies with size of region. (As a
region grows it diversifies, importing less and
so increasing local consumption and the
multiplier (Sirkin 1959)). Also, larger regions
tend to influence neighbors more and so to
enjoy larger feedback effects (Richardson
1972)
WASchaffer
12
Draft 5/11/2010
Review of Economic-Base Literature
Krikelas
APPENDIX A REVIEW OF ECONOMIC-BASE
LITERATURE
The following article appeared in the
Economic Review, Federal Reserve Bank
of Atlanta, July/August 1992, pp. 16-29 ,
and represents the latest in reviews and
critiques of economic-base literature.
the needs of state or local government
officials engaged in the annual
budgeting process, but it would
contribute little information relevant to
long-run local economic development
issues confronting planners and
policymakers. Only rarely is a regional
model able to perform well in more than
one of these distinct decision-making
contexts.1
Why Regions Grow:
A Review of Research
On the EconomicBase Model
The rapid pace of urban growth during
this century, along with the challenge it
has presented for planners trying to
anticipate and influence this growth, has
ensured a healthy demand for regional
economic models, particularly since
1945. Unfortunately, models supplied
have been inadequate.
Andrew C. Krikelas
The author is an economist in the
regional section of the Atlanta Fed's
research department.
Regional economic models are used in a
variety of decision-making contexts.
Government officials use them to
prepare annual budgets. Businesses rely
on them for producing short-run market
demand forecasts and for analyzing
longer-term growth strategies. Urban
planners and transportation officials use
them to develop long-range plans for
urban and regional development. Finally,
state and local policymakers turn to them
to get new ideas for programs and
policies to promote long-run regional
growth.
Although it would be convenient if a
single model had been developed to
serve all these purposes simultaneously,
no such model is ever likely to exist.
Instead, regional models tend to be
highly specialized in terms of the issues
that they are able to address and the time
horizons over which their analytical
results are most reliable. For example, a
short-run forecasting model might serve
Economic Review, FRB Atlanta
At the beginning of the postwar period,
the economic base model was probably
the only such instrument generally
available for regional economic analysis.
This model focuses on regional export
activity as the primary determinant of
local-area growth; it is one of the oldest
and most durable theories of regional
growth, with origins extending at least as
far back as the early 1900s. However,
economic base theory received the
greatest amount of attention from
scholars in regional science between
1950 and 1985. Despite the model's
acceptance over such a long period,
when the noted regional scientist Harry
W. Richardson, writing for a special
twenty-fifth anniversary issue of the
Journal of Regional Science, reflected
upon the more than forty years of
research conducted within this paradigm,
13
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
he concluded that "the findings on
economic base models are conclusive.
The spate of recent research has done
nothing to increase confidence in
them.... The literature would need to be
much more convincing than it has been
hitherto for a disinterested observer to
resist the conclusion that economic base
models should be buried, and without
prospects for resurrection" (1985, 646).
Because regional economic models play
such an important role in planning and
policy discussions, it is important to
have a clear understanding of their
strengths and weaknesses. Limitations of
the economic base model in particular,
because it tends to be widely used,
should be recognized. Recent research
has provided evidence suggesting
substantial improvement in traditionally
static economic base model
specifications through the adoption of
techniques routinely employed in the
macroeconomics time-series literature.
However, this author's research suggests
that these studies may have overstated
the usefulness of these new economic
base model specifications (Andrew C.
Krikelas 1991).
Like Richardson, others over the years
have expressed concern with the narrow
focus of economic base theory on
exports—just one portion of the demand
side of the regional growth equation—to
the exclusion of important supply-side
factors and constraints. Many have
suggested that economic base theory, its
analytical and methodological
techniques, and the public policies that it
promotes should be abandoned in favor
of other, more comprehensive theories of
regional growth and development.
The purpose of this article, therefore, is
twofold. First, a concise analytical
history of the old and extensive
economic base literature generated by a
variety of professional and academic
disciplines is provided in order to place
recent research in perspective. The
discussion then turns to the central
question addressed in Krikelas (1991):
Can techniques borrowed from statistical
time-series literature successfully
breathe new life into the traditional
economic base model?
Nevertheless, economic base research
continues. Most notably, James P.
Lesage and J. David Reed (1989) and
Lesage (1990) have provided empirical
evidence in support of the economic
base hypothesis as both a short-run and
long-run theory of regional growth.
These authors suggest that their models
could be used both for short-term
forecasting of regional employment,
income, and product and for longerrange regional economic planning and
policy analysis. If these claims were
valid, then the economic base model,
rather than being of little value, would
be one of the few regional models that
might be useful in each of these very
different but crucially important
decision-making contexts.
Economic Review, FRB Atlanta
Definition of the Economic Base
Concept
As originally formulated, the economic
base model focused on regional export
activity as the primary source of localarea growth. According to this theory
total economic activity, ET is assumed to
be dichotomous, with a distinction being
made between basic economic activity,
EB (activities devoted to the production
of goods and services ultimately sold to
14
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
consumers outside the region), and
nonbasic economic activity, ENB, which
includes activities involved in producing
goods and services consumed locally:
ET = EB+ ENB
(1)
This division of regional economic
activity into these two distinct sectors is
the central concept of the model.2 A
serious empirical concern is immediately
raised by this approach, however,
because appropriate export data are
available at any subnational level only at
high cost and with long lags. Various
alternative measures have been proposed
and analyzed in the literature over the
years, but none has been found entirely
adequate. Data problems, therefore, have
always complicated economic base
research.
indicates that total economic activity is
primarily a function of basic activity:
ET = + (1 + )* EB
(3)
The expression (1 + ) is commonly
referred to as the economic base
multiplier, and the parameter, , is called
the economic base ratio.
When applied to analyzing regional
growth, the economic base model
suggests that the growth process will be
led by industries that export goods and
services beyond regional boundaries. It
even offers a prediction, captured in the
multiplier, of the total regional impact
likely to result from a change in basic
economic activity generated outside the
region. Understanding the future path of
a regional economy, the model implies,
requires simply concentrating on the
prospects for the base industries. These
few important industries are often
dubbed "engines of regional growth."
While the central concept of the
economic base model is the duality of
regional economic activity, its
fundamental behavioral assumption is
that nonbasic economic activity depends
on basic economic activity. In this
perspective, external demand for a
region's exportable goods and services
injects income into the regional
economy, in turn augmenting local
demand for nonexportable goods and
services. The model assumes that the
income injected into the regional
economy and the accompanying
potential for developing locally oriented,
nonbasic industries are in proportion to
the size of a region's export base. Static
and demand-oriented, the model ignores
factors that affect the supply of a
region's output and other changes, such
as the introduction of new products, that
affect demands.
ENB =f(EB) =  + * EB .
(2)
Equations (1) and (2) can then be
combined into the reduced-form
expression in equation (3), which
Economic Review, FRB Atlanta
This simple model captures the essence
of economic base theory. Although the
model has been enhanced over the years
to include additional variables as well as
to capture more explicitly the dynamic
nature of the regional growth process,
most changes have been made within the
scope of this simple demand-oriented
specification. In general, economic base
models have not evolved to
acknowledge the potential impact of
many important variables that may affect
regional growth—interregional capital
flows; labor migration patterns; changes
in products, tastes, and production
processes; demographic shifts; and
changes in state and local tax laws, to
name a few. Because these issues are
generally too important to ignore, many
regional scientists have concluded that
economic base theory lacks the
complexity to provide a useful
15
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
framework for analyzing many regional
economic issues and policies. The
following review of the development
and testing of the model will summarize
where the debate on this topic stands at
this point.
In 1921 M. Arrousseau made a similar
observation in commenting on the
relationship between what he
distinguished as a town's primary and
secondary occupations: "The primary
occupations are those directly concerned
with the functions of the town. The
secondary occupations are those
concerned with the maintenance of the
well-being of the people engaged in
those of primary nature" (John W.
Alexander 1954, 246).4 Also in 1921,
landscape architect Frederick Law
Olmsted distinguished between what he
called primary and ancillary economic
activity in an urban area (Alexander
1954, 246.)5
History of the Economic Base
Literature
Five fairly distinct chronological periods
characterize the history of the economic
base literature: (1) the origin of the
concept, 1916-21; (2) early
development, 1921-50; (3) the first
round of serious debate, 1950-60; (4) the
second round of debate, 1960-85; and
(5) a third and perhaps final round of
debate begun in 1985 and continuing
today. Decades of research within the
economic base paradigm have created a
body of conventional wisdom
concerning the uses and limitations of
the model, both in theory and in practice.
Nonetheless, as yet another round of
discussion has begun, it seems that few
lessons of the past have been learned and
that a brief summary of the history of
this literature might be useful.
Thus, although Sombart was apparently
the first to observe formally the seeming
duality of urban and regional economic
activity, the remarks of his
contemporaries Arrousseau and Olmsted
make it abundantly clear that the concept
was ripe for expression. By the early
1920s, therefore, the economic base
concept had generally surfaced as a
potential theory for explaining the
regional growth process.
Origin of the Economic Base Concept.
The essential duality of regional
economic activity that is central to the
simple model expressed in the equations
above was first articulated in 1916 by
the German sociologist Werner Sombart,
who wrote of "actual city founders,"
identified as the "active, originative, or
primary city formers"—those whose
positions of authority, wealth, or
occupation allowed them to draw
income from outside the city—and the
"passive or derived or secondary city
founders," whose livelihood depended
on the city formers (Gunter Krumme
1968, 114).3
Economic Review, FRB Atlanta
Early Development of the Theory.
Following establishment of the theory,
the next logical step should have been
the empirical testing of the validity of
the model's central hypothesis. However,
this step was almost universally ignored
and the model adopted as useful as the
rapid growth of cities early in the
century pressured state and local
officials to improve the way in which
they developed plans for urban
expansion and the provision of public
infrastructure and government services.
The economic base model provided a
much-desired framework for developing
such plans, and studies designed to
16
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
identify and measure basic industries—
economic base studies—quickly became
primary tools employed in acquiring
information for long-range planning.
export base industries (Krumme 1968,
113). These calculations placed Berlin's
nonbasic/basic ratio, , at 1.07, an
approximately one-to-one relationship.
Although Sombart did not use this
information to forecast Berlin's growth,
he could have done so. Making a more
limited forecast of the prospects for
those industries he had identified as
being part of the city's export base and
multiplying that total by the city's
economic base multiplier (1 + ) of 2.07
(assuming that the city's base ratio had
remained relatively stable in the
intervening twenty years since the
census was conducted) would have
provided a forecast of the change in total
economic activity expected in Berlin as a
result of some externally generated
change in demand for its export product.
After identifying a region's export base,
economic base studies calculate a localarea economic base ratio, . Once
calculated, the economic base ratio can
be used with forecasts of the future
growth of the region's export base
industries to predict the region's overall
growth. The study's focus on the smaller
number of industries identified as
regional export industries helps
streamline the process of forecasting
total regional economic activity. In
addition, by identifying those industries
considered most important to the
regional growth process, an economic
base study provides information that
adds insight to discussions of regional
industrial policies and programs.
The reliance on secondary data sources
for Sombart's study of Berlin's economic
base is typical of most such research. As
pointed out earlier, even today the
appropriate regional export data required
to conduct an adequate economic base
study are available only at relatively
high cost. The comprehensive economic
analysis of the city of Oskaloosa, Iowa,
published in Fortune magazine in 1938
illustrates this point ("Oskaloosa. . ."
1938).
Although published in a popular
magazine, this study represents an
important contribution to research on the
economic base theory. The magazine
staff conducted a complete census of the
town's 3,000 families in order to
determine the origin and destination of
income flows within the city. They also
conducted a census of the town's
businesses, including an accounting of
the destination of their output and the
source and value of the most important
Sombart's analysis of the Berlin
economy, published in 1927, was the
first economic base study conducted
during this period. Sombart, complaining
that "nobody makes the effort to sit
down with a pencil and figure out with
the help of occupational statistics how
much there actually is of a city-forming
industry in a city such as Berlin,"
developed an empirical approach for
dividing an urban economy into its dual
parts (Krumme 1968,116).6
Lacking detailed information on regional
export activity, Sombart relied upon
industry employment data collected in
Berlin in 1907 to estimate the basic and
nonbasic sectors of the city's economy.
Relying mainly upon his personal
judgment, Sombart estimated that
approximately 262,000 of Berlin's total
work force of 543,000 were employed in
Economic Review, FRB Atlanta
17
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
inputs into the local-area production
process.
economic base idea to be a tool that
might be employed in analyzing the
economic background of cities with the
objective of forecasting the future of the
entire city" (1953a, 163).
The results of the study indicated that in
1937 Oskaloosa was a net exporter of
goods and services to the rest of the
world and that manufactured goods and
professional services were the town's
leading export industries. The study's
findings are interesting because they
were based upon a census that provides a
relatively accurate portrayal of
Oskaloosa's export activity during the
year studied. Even by present standards
this study represents one of the most
thorough economic analyses of a small
community ever published.
In this text Weimer and Hoyt
distinguished between "urban growth"
and "urban service" industries,
suggesting that a region's potential for
growth depended primarily upon the
prospects for the region's urban growth
industries. They provided a six-step
procedure for identifying such
industries. Using relatively accessible
income and employment data, the
authors developed a methodology that
represented a combination of what has
become known as the assignment
technique and the location-quotient
technique of economic base
identification. The assignment technique
is essentially identical to Sombart's
methodology, in which personal
judgment is used to assign industries
within a particular regional economy to
basic and nonbasic sectors. The locationquotient technique, on the other hand,
relies upon regional economic data to
make such distinctions.
The great effort required to collect these
data, however, explains why a survey- or
census-oriented approach to economic
base identification generally has been
abandoned for the nonsurvey
identification techniques made popular
by Homer Hoyt in the late 1930s.
Working with the Federal Housing
Administration during the mid-1930s,
Hoyt developed and employed an
economic base methodology for
producing forecasts of local housing
market demand. His techniques became
known to a wide audience with the
original publication of his textbook,
Principles of Urban Real Estate
(coauthored with Arthur M. Weimer in
1939), which Richard B. Andrews called
the first "complete statement of the
theory of the economic base." In
commenting on the impact of this work,
Andrews continued, "This statement
included much material that was new
outside of technical reports. For
example, it introduced in formal fashion
the idea of a mathematical relation
between basic employment and service
employment.... Hoyt considered the
Economic Review, FRB Atlanta
Location-quotient methodology
compares a region's concentration of
economic activity in a particular industry
with that of a benchmark economy,
usually the entire country in which the
region is located. If the regional
concentration, measured in terms of the
industry's share of total regional
employment or income, exceeds the
benchmark economy's concentration in
that industry, the surplus level of
employment or income is assumed to
measure regional export activity. For
example, if an industry accounts for 6
percent of regional employment but only
18
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
2 percent of national employment, twothirds of that industry's employment
would be called basic. (If the regional
activity in an industry is less than that at
the national level, the industry is
categorized as nonbasic.) Making this
identification requires only industry
employment or income data for the
region and a similar set of data for an
appropriate benchmark economy.
empirical tests were reported during this
entire decade.
The earliest and most cogent critique of
economic base theory was presented by
George Hildebrand and Arthur Mace
(1950) in their analysis of the Los
Angeles metropolitan area. This
important contribution identified the
theoretical model upon which the
economic base paradigm was founded
and performed an empirical test that
provided evidence supporting the
validity of the economic base
hypothesis, at least for short-run
forecasting.
Although Weimer and Hoyt were not the
first to propose using the location
quotient and assignment techniques as
nonsurvey methodologies for dividing
regional economic activity into its basic
and nonbasic components, dissemination
of the techniques through their textbook
introduced these shortcuts to a wide
audience. With these methodologies
available it became feasible for local
development officials to adopt the
economic base paradigm for purposes of
analyzing specific urban and regional
economies. During the latter half of the
1940s, once these techniques had
become more widely known, a much
larger number of cities and states began
to use the economic base model in urban
and regional planning and economic
analysis.7
Hildebrand and Mace's most significant
contribution was their explicit
formulation of economic base theory as
a testable behavioral hypothesis. Their
results, which demonstrated a
statistically significant short-run
relationship between basic and nonbasic
employment in Los Angeles, represented
the first empirical confirmation of the
economic base hypothesis. Furthermore,
the authors formulated their tests within
the context of an explicitly Keynesian
national income model and then outlined
the inherent limitations of such a model.
Theoretical Debate. By 1950 economic
base theory and its methodological
techniques had become established as
the primary tools of regional planning.
The theory itself had been accepted,
uncritically, as an explanation of localarea growth and economic development.
Between 1950 and 1960, however,
discussion at the theoretical and
methodological level turned directly to
the question of the validity of the
economic base hypothesis itself.
Unfortunately, only a handful of
Economic Review, FRB Atlanta
Consider the familiar Keynesian
relationship:
Y = C + I + G + (X - M),
(4)
where total regional income, Y, is
divided into a number of distinct sectors,
including consumption, C; investment, I;
government expenditures, G; and
exports minus imports, X - M. The
reduced-form expression of this model
would include some smaller set of
exogenous variables, only one of which
would be regional exports. (Other
exogenous variables would include the
autonomous components of
19
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
consumption, investment, government
expenditures, and imports; marginal
propensities to consume locally, to
invest locally, and to import; and local
and federal tax policies.) It is this set of
exogenous factors that would determine,
theoretically, a region's total income
level, Y.
Unfortunately, the lessons contained in
Hildebrand and Mace's study were not
widely disseminated. Hildebrand and
Mace were among the first economists to
contribute to the economic base
literature. Their article was published in
a journal not normally read by
geographers and urban planners, who,
before 1950, had played a dominant role
in the research conducted within the
economic base paradigm. Therefore,
rather than playing the role of a seminal
article to a further body of empirical
research, the Hildebrand and Mace
article remained relatively unknown.
The debate of the 1950s brought many
of their important insights to the
attention of geographers and urban
planners, but it took nearly a decade for
all of these contributions to be
uncovered.
The economic base model focuses on
one particular aspect of this relationship,
regional export activity, X (EB in
equation [1] above), and can be
considered a special case of the more
general Keynesian model in equation
(4). Given this interpretation, it becomes
clear that for exports to be considered
the only exogenous determinant of
regional growth, all other relevant
factors, related to both demand and
supply, must remain fairly constant or be
functions of export activity. Although
this might be a tenable assumption in the
short run, it probably is an extremely
poor one in the long run. Hildebrand and
Mace made this observation explicit and
suggested that the model was most
appropriate for anticipating regional
economic trends over a short time
horizon. In addition, they listed some of
the other variables that they thought
should be taken into account in
developing a more comprehensive model
of regional economic activity:
population levels and interregional
migration patterns, regional capital
investment levels and annual flows, state
and local tax policies, and changes in the
cost of transportation to reach external
markets. Despite these reservations,
Hildebrand and Mace offered a fairly
encouraging assessment of the prospects
for this type of research, based on the
availability of additional census data and
further empirical analysis across a tenyear span. 8
Economic Review, FRB Atlanta
Most of the 1950s' debate on economic
base theory was conducted in the
geography and planning literatures. The
origin of this debate can be traced to a
series of nine articles published by
Andrews between 1953 and 1956 (see
reference list). These articles provided a
careful exposition of economic base
theory and the methodologies that had
been developed to analyze urban and
regional economic activity. The author's
stated purpose was to explore and
evaluate the entire concept. "We have
operated far too long on a set of ideas
which appear valid but which, despite
substantial conceptual omissions and
difficulties of application, seem to be
accepted all too blithely," he wrote,
calling for "more fundamental thinking
on and questioning of the reality and
utility of base theory as presently
conceived" (1953a, 167).
20
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
While Andrews was somewhat critical in
his assessment of the economic base
paradigm, he clearly was a proponent of
its inherent validity and usefulness.
Instead of suggesting the abandonment
of the model as a tool for urban and
regional economic analysis, he identified
ways in which it could be improved to
serve such purposes better. His
recommendation included better efforts
at basic industry identification and
measurement, improvements in the
collection of regional data, and
modifications in the way in which
economic base concepts were used.
Given Andrews's criticism of the state of
the economic base research prior to
1950, it is surprising to note he did not
address one of the most fundamental
shortcomings of this research: the lack of
empirical verification of the underlying
hypothesis. Krikelas (1991) identified
only five empirical tests of the economic
base hypothesis conducted during the
1950s. Three of those studies, including
that of Hildebrand and Mace, supported
the validity of the economic base
hypothesis, at least in the short run, and
two provided evidence against it. A
decade of research, therefore, provided
insufficient empirical evidence for
determining the validity of the model's
central hypothesis.
and practice rather than testing. Hans
Blumenfeld (1955) was critical of the
economic base model's narrow focus on
export activity as the primary source of
regional growth. While he agreed that
this model might do well to explain
economic growth in small or highly
specialized economies, he argued that it
was inadequate to explain the growth of
complex urban economies. Blumenfeld
was also critical of the policy
implications of the model; these focused
almost exclusively on supporting
existing export industries at the expense
of other reasonable alternatives, such as
fostering the establishment and
development of industries that would
compete with imported goods and
services.
Charles M. Tiebout (1956a, 1956b) and
Douglass C. North (1955, 1956) engaged
in a short but lively debate over the
short-run versus long-run applicability of
the economic base model. Tiebout,
explicitly recognizing the Keynesian
roots of the economic base model,
supported Hildebrand and Mace's (1950)
contention that the economic base model
was most appropriate for short-run
economic analysis. He also argued that
the economic base model minimized the
important contribution that nonbasic
economic activity made to local area
growth and development. He wrote that,
although export activity was important,
"in terms of causation, the nature of the
residentiary industries will be a key
factor in any possible development.
Without the ability to develop
residentiary activities, the cost of
development of export activities will be
prohibitive" (1956a, 164).
When applied to analyzing
regional growth, the economic
base model suggests that the
growth process will be led by
industries that export goods and
services beyond regional
boundaries.
North, however, objected to the
characterization of the economic base
Instead, most of the debate of the 1950s
centered on questions related to theory
Economic Review, FRB Atlanta
21
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
model as an adaptation of the demandoriented Keynesian model. Instead, he
argued that the most important
determinant of a region's long-run
growth potential was its ability to attract
capital and labor into the region from
outside. Such supply-enhancing flows in
turn would respond quite favorably to
profit opportunities offered by regions
engaged in high levels of export activity.
North observed that historically "it was
frequently the opportunities in
manufacturing for the United States
market which led to immigration of
labor and capital into a region. The
important point is that the pull of
economic opportunity as a result of a
comparative advantage in producing
goods and services in demand in existing
markets was the principal factor in the
differential rates of growth of regions"
(1956, 166).
regional growth, both ultimately agreed
that supply factors needed to be added to
the model in order to make it relevant for
long-run regional economic analysis.
One additional advance in the theoretical
literature of this period that called into
question the adequacy of economic base
modeling techniques was the
development of regional input-output
models. Before 1950 the economic base
model represented the primary tool
available to regional planners for
analyzing the impacts of anticipated
changes in regional economic activity.
During the first half of the 1950s,
however, input-output modeling
techniques first developed by Wassily
W. Leontief (1951) were adapted for
purposes of regional economic analysis.9
While a regional input-output model
could distinguish between the
differential regional impacts that might
be associated with, for example, the
construction of a specialty steel
manufacturer versus a mail-order catalog
facility—two very different kinds of
basic economic activity—the simple
two-sector economic base model could
not make such a distinction. Given this
limitation, many urban planners began to
advocate input-output techniques as
more appropriate for forecasting
anticipated changes in regional
economic activity.
Many regional scientists have
concluded that economic base
theory lacks the complexity to
provide a useful framework for
analyzing many regional economic
issues and policies.
The economic base model proposed by
North explicitly recognized the
important role of supply factors in
determining the nature and growth
potential of a region's export base. In
practice, however, most economic base
models of this and subsequent periods
have maintained a fairly strict demand
orientation. This demand-oriented model
is also the one to which Tiebout raised
so many objections. As a result, although
Tiebout and North found themselves on
different sides concerning the validity of
the model as a long-run theory of
Economic Review, FRB Atlanta
The debate of the 1950s also focused on
several important methodological issues.
Papers by John M. Mattila and Wilbur
R. Thompson (1955) and Charles L.
Leven (1956) considered the adequacy
of the location-quotient technique's
ability to identify a region's economic
base industries. While suggesting certain
improvements to the traditional
formulation of the location quotient,
22
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
Mattila and Thompson concluded that "if
used with care, the index of surplus
workers in both its absolute and relative
form should prove to be a highly useful
tool in regional economic base studies"
(1955, 227).10 Leven, on the other hand,
arrived at the opposite conclusion,
stating that "the shortcomings of this
technique render it useless as a
quantitative measure of basic activity in
an area" (1956, 256).
economies, especially at the substate
level, except with long lags.
By the beginning of the 1960s
professionals engaged in urban and
regional economic analysis had divided
into three distinct camps concerning the
conduct of research within the economic
base paradigm: those who still
considered the economic base model to
be a reasonable framework for urban and
regional economic analysis; those who
questioned its validity but sought more
empirical evidence before abandoning
the paradigm; and those who rejected the
validity of the hypothesis, instead
turning to the investigation of other
methods of regional economic analysis,
including regional input-output models.
Whereas the debate of the 1950s was
conducted primarily at the theoretical
level, the quarter-century between 1960
and 1985 was filled with empirical
examinations of a wide range of
theoretical and methodological questions
related to the economic base model.
The issue of the appropriate measure to
be used for calculating location quotients
was also discussed. Because
employment data were more readily
available than wage or income data,
most economic base studies of this
period used employment in identifying
regional export activity. This measure,
however, has some serious drawbacks.
In addition to placing equal weight upon
part-time and full-time employment and
failing to adjust adequately for
productivity and wage differences
between workers employed in different
industries, employment data do not
provide any measure of the impact that
transfer payments and other sources of
unearned income, such as interest
payments, rents, and profits, have upon a
regional economy.
Empirical Debate. Between 1960 and
1985 a large number of articles and
several books were published on the
economic base model.11 Yet while the
question of the empirical relevance of
the economic base hypothesis was
arguably the most important issue facing
the profession on the heels of the debate
of the 1950s, only a quarter of these
contributions actually addressed it.
To provide some perspective on the
extensive literature of this period,
Krikelas (1991) developed a taxonomy.
The six categories listed represent
distinct facets of the economic base
literature of this period: (1) identification
of export base activity, (2) calibration
studies, (3) extensions of the base model,
(4) case studies, (5) theoretical works,
Recognizing the serious weaknesses
associated with the use of employment
data for purposes of identifying a
region's economic base, Andrews
(1954a), Leven (1956) and Tiebout
(1956c) all suggested the adoption of
alternative measures of regional
economic activity. Andrews and Tiebout
advocated the use of income received by
residents of the region, and Leven
argued for a value-added measure.
Income and value-added data, however,
generally are not available for regional
Economic Review, FRB Atlanta
23
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
and (6) tests of the economic base
hypothesis.
nonsurvey techniques. Another
seventeen studies conducted between
1960 and 1985 can be classified as
calibration studies, and Isserman
provides an excellent summary of such
research, concluding that although
efforts to develop and refine the
nonsurvey methods had been substantial,
"the situation is lamentable" (1980, 17979).
A thorough discussion of the
contributions that fall into each of these
categories is beyond the scope of this
article. However, a summary of the
major developments in each category
should yield insights. It should be noted
that the majority of the research
published during this period—that is,
categories (l)-(4)—assumed, at least
implicitly, the validity of the economic
base hypothesis.
Extensions of the Base Model. During
this period at least two important
extensions were made to the simple
economic base model. In the first,
additional variables other than basic
economic activity were added to the
original specification in order to
investigate their effects on the regional
growth process. Stanislaw Czamanski's
(1965) study represents the first of
several in which a demographic
variable—population—was explicitly
included in the model specification. Paul
E. Polzin (1977), on the other hand,
developed a model designed to capture
the effects of local-area labor supply
conditions on regional economic
activity, and Ron E. Shaffer (1983) and
Shahin Shahidsaless, William Gillis, and
Shaffer (1983) included variables
designed to measure the contribution of
both demographic and geographic
factors. Given the fact that these authors
generally found the additional variables
to be very important determinants of
regional growth, it is somewhat
surprising that relatively few studies
focused on this issue.
Identification of Export Base Activity.
The most contentious issue facing
researchers using the economic base
model is the identification of regional
export activity. Much attention has been
paid to the development of nonsurvey
techniques, and during this period
seventeen studies were devoted to
creating new or improving old
methodologies. Edward L. Ullman and
Michael F. Dacey (1960) and Vijay K.
Mathur and Harvey S. Rosen (1974)
introduced two completely new
nonsurvey methods for identifying
regional export activity, and several
other researchers suggested refinements
for improving both the location-quotient
and assignment methods of economic
base identification. Andrew M. Isserman
(1980) offers an excellent survey of the
developments of this period, including a
critique of each methodology.
Calibration Studies. Calibration studies
are research designed to test the
adequacy of competing nonsurvey
identification techniques. Researchers
either compare nonsurvey estimates of
regional exports with benchmark survey
or census data on regional exports or
simply compare results of several
Economic Review, FRB Atlanta
A second innovation, which gained a
much broader acceptance in the
literature, was the disaggregation of
basic activity into more than one
sector—manufacturing, construction,
services, and government, for example.
24
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
This work was stimulated by the
challenge posed by regional input-output
models and their clear demonstration
that changes in regional activity in
different export industries were likely to
have very different effects upon a
regional economy. Steven J. Weiss and
Edwin C. Gooding (1968) provide the
first example of a multisectoral
economic base model, and their work
was repeated and extended in many
subsequent studies. However, while the
literature of this period reported the
results of numerous multisectoral
economic base models, the maximum
number of sectors for which multipliers
can be estimated has always been limited
by the length of available data series,
usually to ten sectors or fewer. As a
result, no economic base model has ever
been able to reproduce the level of
industry disaggregation available in most
regional input-output models.
the other hand, demonstrated the close
relationship between economic base and
international trade models. And finally,
Wolfgang Mayer and Saul Pleeter
(1975) and F.J.B. Stillwell and B.D.
Boatwright (1971) developed economic
base theoretic models that demonstrated
that the location-quotient and minimumrequirements methods of export industry
identification could be derived from, and
were consistent with, economic base
theory. While these and other
contributions provided a formal
statement of the theoretical
underpinnings of the economic base
model and its methodological
techniques, they did not provide
empirical evidence in support of the
theory's central hypothesis.
Tests of the Economic Base Hypothesis.
In considering the empirical results of
studies published during this period, it is
important to distinguish between
dynamic and static tests of the economic
base hypothesis. Although the economic
base paradigm generally has been used,
implicitly, to analyze dynamic regional
economic events, most specifications of
the model, like that in equations (l)-(4),
have been explicitly static in nature. This
point was made clear first by Charles E.
Ferguson (1960). Subsequently, one of
the major contributions of this period
was the more explicit consideration of
the dynamic properties of the economic
base model. Researchers began using
time-series modeling and other
econometric techniques to analyze the
short-run versus long-run applicability of
the economic base model as well as to
develop practical regional forecasting
models.
Case Studies. In most instances the main
purpose of these base studies was the
calculation of multisectoral economic
base multipliers intended to demonstrate
the significant impact of the sectors
under consideration. Early studies had
focused mainly on the role of
manufacturing in the regional growth
process. Many of these later works were
instead devoted to showing the
important contribution that the trade and
service sectors could also play in
regional growth.12
Theoretical Works. Several contributions
during this period were devoted
exclusively to advancing the theoretical
foundations of the economic base
paradigm. Edwin F. Terry (1965)
explicitly derived the linkage between
the economic base model and the
Keynesian model. John Mutti (1981), on
Economic Review, FRB Atlanta
The majority of these studies, however,
were still predicated upon explicitly
25
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
static model specifications. Even some
of the studies that ostensibly attempted
to capture the dynamic properties of the
economic base model failed to do so
adequately.13 Given that the utility of an
economic base study depends upon its
use for analyzing dynamic economic
events, it is unfortunate—and
surprising—that relatively few of these
empirical studies were specified in such
a way as to explore this issue.
existence of a long-run economic
relationship between basic and nonbasic
employment. Thus, while a narrow
majority of the test results reported
during this twenty-five-year period
provided evidence in support of the
validity of the economic base
hypothesis, at least in the short run, very
little empirical evidence suggested that
the model could also perform well in the
long run.
In reviewing the literature of this period,
Krikelas (1991) examined twenty-three
studies that reported the results of tests
of the economic base hypothesis. Eleven
were static tests; twelve, dynamic. Of
these, six static tests and seven dynamic
tests provided results consistent with the
economic base hypothesis. Many of the
dynamic tests of the hypothesis were
further designed to explore the issue of
the short-run versus long-run validity of
the economic base hypothesis. Only four
studies—Harold T. Moody and Frank
W. Puffer (1970), Curtis Braschler
(1972), Braschler and John A. Kuehn
(1975), and James E. McNulty (I977)—
provided any ostensible evidence in
support of economic base theory as a
long-run theory of regional growth.
By 1985 the most definite and positive
comment the literature could support
about an economic base model was that
it would perform best in providing
relatively short-term forecasts of total
regional economic activity. More than
fifty years of research had failed to
provide any substantial evidence in
support of the model as a long-run
theory of regional growth—a serious
limitation in light of the fact that
policymakers are generally more
interested in long-run growth issues. It
should be clear that the economic base
model, because it fails to account for
some of the fundamental determinants of
the regional growth process, should not
be adopted for long-range planning and
policy analysis. These are the results that
led to Richardson's call (cited earlier) for
burying economic base models "without
prospects for resurrection" (1985, 646).
As Shelby D. Gerking and Isserman
(1981) have pointed out, however, the
model specifications adopted in three of
these four studies actually tested only the
contemporaneous relationship between
basic and nonbasic economic activity
rather than the long-run relationship
purportedly tested by the authors. They
further concluded that Moody and
Puffer's (1970) results, which were
based upon an appropriately specified
dynamic model, were more likely to be
attributable to the authors' choice of
bifurcation methodology than to the
Economic Review, FRB Atlanta
Third Period of Debate. Despite
Richardson's impassioned warning,
research continues to be performed
within the framework of the economic
base paradigm. Recently, a resurgence in
such research has been fueled by a
recognition that some sophisticated
econometric techniques used in analysis
of macroeconomics time series may be
applied to the economic base model. In
particular, it has been demonstrated that
26
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
the essential features of the economic
base model can be captured within the
context of a bivariate vector
autoregression (VAR) linking basic and
nonbasic economic activity.14 Once
specified, such a VAR can be subjected
to the time-series econometric tests and
analytical procedures that have been
developed over the years. Granger
causality tests can be formulated in order
to test the validity of the economic base
hypothesis. Impulse-response functions
(the response of a variable to an
unanticipated increase in other variables)
can be derived and given a natural
interpretation as dynamic base
multipliers. Forecasting competitions
can be held in order to assess how well
competing models improve the accuracy
of a given forecast. Finally, cointegration tests can be performed in
order to assess whether there might be a
long-run relationship between basic and
nonbasic economic activity.
integration tests that demonstrated a
long-run economic relationship between
basic and nonbasic employment in
several of these MSAs, the combined
results of this research effort seemed to
provide evidence that such empirical
work was both justified and could prove
fruitful.
A third period of debate on the
economic base model centers on
the question of whether new
techniques borrowed from
macroeconomics time-series
literature can revive the
traditional economic base model.
The results of Lesage and Reed's (1989)
and Lesage's (1990) studies are already
being cited in the literature. David S.
Kraybill and Jeffrey Dorfman (I992), for
example, used these authors'
methodology to estimate a three-sector
model for the state of Georgia. These
and other recent contributions represent
examples of what has become a third
period of debate on the economic base
model, centered on the question of
whether new techniques borrowed from
macroeconomics time-series literature
can revive the traditional economic base
model.
Using such techniques, Lesage and Reed
(1989) and Lesage (1990) found
empirical evidence in support of the
economic base hypothesis. Lesage and
Reed reported Granger causality test
results that were generally consistent
with the economic base hypothesis, at
least in the short run. Proceeding further,
the authors used their VAR model
specifications to derive impulseresponse functions describing the
dynamic relationships between basic and
nonbasic employment in eight
metropolitan statistical areas (MSAs) in
Ohio. The reasonable nature of the
multipliers calculated from this
experiment led the authors to conclude
that this methodology offered promise
for regional economic forecasting and
policy analysis purposes. When Lesage
(1990) reported the results of coEconomic Review, FRB Atlanta
Replicating and expanding this research,
this author conducted extensive timeseries econometric tests of the economic
base hypothesis on models specified for
the state of Wisconsin (Krikelas 1991).
The results of this research, based upon a
large number of two-sector and
multisector model specifications, suggest
that these new techniques do not provide
the convincing evidence to support
revival of the economic base model for
27
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
purposes of long-term forecasting or
planning context.
multiplier estimates becomes so arbitrary
as to call into question the credibility of
the multipliers derived from such
specifications. As a result, any policy
implications that might be implicit in a
finding of significant differences
between sectoral multiplier estimates
would also be questionable.
First and foremost, the fundamental
problems associated with deriving
adequate estimates of regional export
activity remain unresolved. Although
Lesage and Reed (1989) claimed that
their dynamic location-quotient
technique "provides a more accurate
decomposition of local area
employment" (1989, 616), this claim
seems to be overstated. Krikelas (1991)
confirms the results reported by
Isserman (1980) and several others who
have found that the location-quotient
technique tends to underestimate the
level of regional export activity and,
consequently, lend an upward bias to
export base multiplier estimates.
More fundamentally, however, Krikelas
concludes that the new techniques
employed in Lesage and Reed and
similar research do nothing to broaden
the economic base paradigm's focus on
the demand side of the regional growth
equation. Past research has clearly
indicated that economic base models that
fail to account for important supply-side
factors and constraints do not perform as
well as models that try to incorporate
such relationships. Labor migration
patterns, interregional capital flows, and
state and local tax policies all have
important effects upon regional
economic growth and development and
need to be incorporated into regional
economic model specifications for the
model to have value for anything other
than short-term forecasting. Although it
is possible to expand the bivariate
economic base VAR to include some of
these important supply-side variables,
this author has concluded that such
research would be largely in vain
because other problems would remain
(see Krikelas 1991). The recent attempt
to breathe new life into the economic
base model seems to have failed to
resuscitate the patient.
Second, in order to assess the stability of
multiplier estimates derived from a
bivariate VAR, Krikelas (1991
calculated impulse response functions
for models that were based upon data
generated from a variety of alternative
sample separation techniques. The
results of this experiment show that
small changes in the way in which a
given data set is divided into its basic
and nonbasic components can lead to
large changes in multiplier estimates.
These results call into question the
usefulness of the dynamic multipliers
derived from a bivariate economic base
VAR for even short-run regional impact
analysis.
Finally, Krikelas (1991) explored the
possibility of deriving multipliers from
multisectoral VAR specifications and
found similar difficulties. As the number
of sectors included in a VAR is
expanded, establishing identifying
restrictions required in order to derive
Economic Review, FRB Atlanta
Conclusion
Given the fact that several authors have
begun to report empirical results in
support of the validity of the economic
base hypothesis, a third round of debate
on the model seems already under way
28
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
in the literature. An examination of some
of the claims made by the proponents of
these new dynamic economic base
models, however, indicates that they are
apparently unaware of the scope of the
literature preceding their efforts.15 This
brief analytical history should be
sufficient to convince users that the
economic base model has severe
limitations, especially for economic
planning and policy analysis, and to help
make this next and perhaps final round
of debate a relatively short-lived one.
Notes
1.
Structural econometric models are often
used for purposes of both forecasting and
policy analysis. However, the great
expense required to specify and maintain
such models has generally led economists
either to develop less complex models that
focus narrowly on a small set of policy
issues or to develop theoretical time-series
models that perform well for purposes of
short-run economic forecasting.
2.
Besides the terms basic and nonbasic, a
number of others have been proposed to
distinguish between the two types of
economic activity: town builders/town
fillers, exchange production/own
production, primary/ancillary, export/local,
as well as others. Andrews (1953b)
directly addresses the issue of the profligate
and confusing terminology of the economic
base paradigm.
3.
their productive occupations may be
roughly divided into those which can be
considered primary, such as carrying on the
marine shipping business of the port and
manufacturing goods for general use (i.e.,
not confined to use within the community
itself), and those occupations which may be
called ancillary, such as are devoted
directly or indirectly to the service and
convenience of the people engaged in the
primary occupations."
Krumme was translating Werner Sombart's
Der Moderne Kapitalismus, Erster Band:
Die Vorkapitalistische Wirtschaft, 2nd rev.
ed. (Munich: Duncker and Humblot, 1916).
Sombart identified the city formers as "a
king who collects taxes; a landlord who
receives rent payments; a merchant who
profits from trade with outsiders; a
craftsman, a manufacturer, who sells
industrial products to the outside; an
author, whose writings are being bought
outside the gates; a physician, who has
clients in the countryside; a student, who is
supported by his parents in another place,
etc. These are the people who live and let
live."
4.
Alexander was citing M. Arrousseau, "The
Distribution of Population: A Constructive
Problem," Geographical Review 11 (1921).
5.
Alexander cites a letter dated February 21,
1921, to John M. Glenn, a member of the
New York Regional Planning Committee in
which Olmsted wrote, "The multiplicity of
Economic Review, FRB Atlanta
29
6.
According to Krumme's translation,
Sombart wrote, "It is necessary to find out
for each trade how much of it is engaged in
work for local consumption and how much
in work for exports out of the city. This
figure then is the city-forming ratio for the
individual trade. Naturally, the ratio can be
found accurately only with the assistance of
an extensive enquete (survey). However,
one could gain at least an approximate
impression of the shares of the export
industries in the total gainful employment
by a careful investigation of the results of
the occupational census" (1968, 116). The
empirical study cited by Krumme was
published for the first time in the second
revised edition of Sombart's Der Moderne
Kapitalismus, Drifter Band: Das
Wirtschaftsleben im Zeitalterr des
Hochkapitalismus, in 1927. Krumme,
however, was quoting from the third
printing of this edition, published in Berlin
in 1955.
7.
The following list identifies a few of the
communities that performed economic base
studies during the 1940s, the individuals or
institutions that performed these analyses,
and the base ratios (,B) calculated,
respectively: New York, The Regional Plan
Association Inc., 2.1; Detroit, Detroit City
Plan Commission, 1.1; Cincinnati, Victor
Roterus and the staff of Cincinnati City
Planning Commission, 1.7; Washington,
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
D.C., National Capitol Park and Planning,
1.1; Brockton, Massachusetts, Homer Hoyt,
0.8; the state of New Jersey, Homer Hoyt,
1. 1; and Albuquerque, New Mexico,
Federal Reserve Bank of Kansas City, 0.9.
This information was originally compiled
by Edward Ullman and published in the
third edition of Weimer and Hoyt's text in
1954 and was reprinted in Pfouts (1960,
30).
8.
9.
which they were calculated, include the
following, respectively: retail trade
multipliers calculated by Friedly (1965) for
Redondo Beach, California; trade and
service sector multipliers calculated by
Terry (1965) for St. Louis, Missouri;
defense industry multipliers calculated by
Billings (1970) for the state of Arizona and
by Erickson (1977) for the Badger
Ammunition Plant, near Baraboo,
Wisconsin; rural area multipliers calculated
by Garrison (1972) for five
nonmetropolitan counties in Kansas; and
university sector multipliers calculated by
Wilson (1977) for Tulsa, Oklahoma.
Hildebrand and Mace wrote, "The
forthcoming Census of 1950 will permit
further advances in this research.
Recalculation of location quotients and
comparisons with 1940 will indicate
changes in external markets and locational
concentrations during the war decade,
particularly in communities undergoing
large gains or losses in population. With
monthly statistics of insured employment, a
current record of employment in nonlocalized industries can be maintained.
Improved multiplier analysis, with current
local labor force statistics, should then
permit more precise depiction of local
unemployment problems, and attainment of
more adequate policies at the o r-all and
community levels" (1950, 249).
13. See Gerking and Isserman's (1981)
discussion of the results of Braschler
(1972), Braschler and Kuehn (1975), and
McNulty (1977).
14. A VAR model consists of an equation for
each variable in which the equations are
estimated by regressing each of the
variables against lagged values of all the
variables. By not imposing any particular
theoretical connection among the variables,
the VAR will capture any correlations that
exist in the data. In this sense, VARs are
distinct from traditional structural models,
which typically include a large number of
variables that are theoretically linked.
Perhaps the most often-cited contribution to
the early regional input-output literature
was an article coauthored by Isard and
Kuenne (I 95 3).
15.
10. The index of surplus workers is simply a
measure of the number of workers in
excess of that which would be required if
the region's employment profile matched
the national average.
11. Krikelas (1991) identified eighty-four
contributions to the literature during this
period.
12. Some of the sectoral multiplier studies
conducted and the region or project for
Lesage, for example, reported on one of
the few empirical tests recorded in the
history of the literature that supports the
economic base hypothesis as a long-run
theory of regional growth and wrote that
"this finding would not be particularly
surprising to most regional economists"
(1990, 309). His is one of several
comments published recently that have
pointed toward the need for presentation of
a comprehensive history of the extensive
body of literature that exists.
References
Alexander, John W.,. "The Basic-Nonbasic Concept of Urban Economic Functions." Economic Geography
30 (1954): 246-6 1.
Andrews, Richard B., "Mechanics of the Urban Economic Base: Historical Development of the Base
Concept." Land Economics 29 (1953a): 161-67.
______. "The Problem of Terminology." Land Economics 29 (1953b): 263-68.
.______. "A Classification of Base Types." Land Economics 29 (1953c):343-49.
______. "The Problem of Base Measurement." Land Economics 30 (1954a): 52-60.
Economic Review, FRB Atlanta
30
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
______. "General Problems of Base Identification." Land Economics 30 (1954b): 164-72.
______. "Special Problems of Base Identification." Land Economics 30 (1954c): 260-69.
______. "The Problem of Base Area Delimitation." Land Economics 30 (1954d): 309-19.
______. "The Concept of The Base Ratios." Land Economics 31 (1955): 47-53.
______. "The Base Concept and the Planning Process." Land Economics 32 (1956): 69-84.
Billings, R. Bruce, "Regional Defense Impact-A Case Study Comparison of Measurement Techniques."
Journal of Regional Science 10 (1970): 199-216.
Blumenfeld, Hans, "The Economic Base of the Metropolis." Journal of the American Institute of Planners
21 (1955): 114-32.
Braschler, Curtis,. "A Comparison of Least-Squares Estimates of Regional Employment Multipliers with
Other Methods." Journal of Regional Science 12 (1972): 457-68.
______., and John A. Kuehn. "Industry Sectors and the Export Base Determination of Nonmetropolitan
Employment Change in Four Midwestern States." Review of Regional Studies 3 (1975): 82-89.
Czamanski, Stanislaw, "A Method of Forecasting Metropolitan Growth by Means of Distributed Lags
Analysis." Journal of Regional Science 6 (1965): 35-49.
Erickson, Rodney A., "Sub-Regional Impact Multipliers: Income Spread Effects from a Major Defense
Installation." Economic Geography 53 (1977): 283-94.
Ferguson, Charles E., "Statics, Dynamics, and the Economic Base." In Techniques of Urban Economic
Analysis, edited by Ralph W. Pfouts, 325-39. West Trenton, N.J.: Chandler and Davis Publishing
Company, 1960.
Friedly, Philip., "A Note on the Retail Trade Multiplier and Residential Mobility." Journal of Regional
Science 6 (1965): 57-63.
Garrison, Charles B., "The Impact of New Industry: An Application of the Economic Base Multiplier to
Small Rural Areas." Land Economics 48 (1972): 329-37.
Gerking, Shelby D., and Andrew M. Isserman, "Bifurcation and the Time Pattern of Impacts in the
Economic Base Model." Journal of Regional Science 21 (1981): 451-67.
Hildebrand, George, and Arthur Mace, "The Employment Multiplier in an Expanding Industrial Market:
Los Angeles County, 1940-47." Review of Economics and Statistics 32 (1950): 241-49.
Isard, Walter, and Robert Kuenne, "The Impact of Steel upon the Greater New York-Philadelphia Industrial
Region." Review of Economics and Statistics 35 (1953): 289-30 1.
Isserman, Andrew M., "Estimating Export Activity in a Regional Economy: A Theoretical and Empirical
Analysis of Alternative Methods." International Regional Science Review 5 (1980): 155-84.
Kraybill, David S., and Jeffrey Dorfman,. "A Dynamic Intersectoral Model of Regional Economic
Growth." Journal of Regional Science 32 (1992): 1-17.
Krikelas, Andrew C., "Industry Structure and Regional Growth: A Vector Autoregression Forecasting
Model of the Wisconsin Regional Economy." Ph.D. diss., University of Wisconsin-Madison, 1991.
Krumme, Gunter,. "Werner Sombart and the Economic Base Concept." Land Economics 48 (1968): 11216.
Leontief, Wassily W. The Structure, of the American Economy. New York: Oxford University Press, 195 1.
Lesage, James P. "Forecasting Metropolitan Employment Using an Export-Base Error-Correction Model."
Journal of Regional Science 30 (1990): 307-23.
______., and J. David Reed. "The Dynamic Relationship between Export, Local, and Total Area
Employment." Regional Science and Urban Economics 19 (1989): 615-36.
Economic Review, FRB Atlanta
31
Jul/Aug 1992
Review of Economic-Base Literature
Krikelas
Leven, Charles L. "Measuring the Economic Base." Papers and Proceedings of the Regional Science
Association 2 (1956): 250-58.
Mathur, Vijay K., and Harvey S. Rosen. "Regional Employment Multiplier: A New Approach." Land
Economics 50 (1974): 93-96.
Mattila, John M., and Wilbur R. Thompson. "The Measurement of the Economic Base of the Metropolitan
Area." Land Economics 31 (1955): 215-28.
Mayer, Wolfgang, and Saul Pleeter. "A Theoretical Justification for the Use of Location Quotients."
Regional Science and Urban Economics 5 (1975): 343-55.
McNulty, James E. "A Test of the Time Dimension in Economic Base Analysis." Land Economics 53
(1977): 359-68.
Moody, Harold T., and Frank W. Puffer. "The Empirical Verification of the Urban Base Multiplier:
Traditional and Adjustment Process Models." Land Economics 46 (1970): 91-98.
Mutti, John. "Regional Analysis from the Standpoint of International Trade: Is It a Useful Perspective?"
International Regional Science Review 6 (1981): 95-120.
North, Douglass C. "Location Theory and Regional Economic Growth." Journal of Political Economy 63
(1955): 243-58.
______. "A Reply." Journal of Political Economy 64 (1956): 165-68.
"Oskaloosa vs. The United States." Fortune, April 1938, 58ff. Pfouts, Ralph W., ed. Techniques of Urban
Economic Analysis. West Trenton, N.J.: Chandler and Davis Publishing Company, 1960.
Polzin, Paul E. "Urban Labor Markets: A Two-Sector Approach." Growth and Change 8 (1977): 11-15..
Richardson, Harry W. "Input-Output and Economic Base Multipliers: Looking Backward and Forward."
Journal of Regional Science 25, no. 4 (1985): 607-61.
Shaffer, Ron E. "A Test of the Differences in Export Base Multipliers in Expanding and Contracting
Economies." Regional Science Perspectives 13 (1983): 61-74.
Shahidsaless, Shahin, William Gillis, and Ron Shaffer. "Community Characteristics and Employment
Multipliers in Nonmetropolitan Counties, 1950-70." Land Economics 59 (1983): 84-93.
Stillwell, F.J.B., and B.D. Boatwright. "A Method of Estimating Interregional Trade Flows." Regional and
Urban Economics 1 (1971): 77-87.
Terry, Edwin F. "Linear Estimators of the Export Employment Multiplier." Journal of Regional Science 6
(1965): 17-34.
Tiebout, Charles M. "Exports and Regional Economic Growth." Journal of Political Economy 64 (1956a):
160-64.
______. "Rejoinder." Journal of Political Economy 64 (1956b): 169.
______. "The Urban Economic Base Reconsidered." Land Economics 32 (1956c): 95-99.
Ullman, Edward L., and Michael F. Dacey. "The Minimum Requirements Approach to the Urban
Economic Base." Papers and Proceedings of the Regional Science Association 6 (1960): 175-94.
Weimer, Arthur M., and Homer Hoyt. Principles of Real Estate. Rev. ed. New York: Roland Press
Company, 1948.
Weiss, Steven J., and Edwin C. Gooding. "Estimation of Differential Employment Multipliers in a Small
Regional Economy." Land Economics 44 (1968): 235-44.
Wilson, J. Holton. "Impact Analysis and Multiplier Specification." Growth and Change 8 (1977): 42-4.
Economic Review, FRB Atlanta
32
Jul/Aug 1992
Techniques for Analysis
Note A
APPENDIX B AN ECONOMIC-BASE MODEL
OF ATLANTA
The following article is an example of a simple economic-base multiplier, probably the
last computed for Atlanta. With modern data sources and computing equipment, such
models are seldom seen; but just a few decades ago they were quite common. Their
virtue is the ease with which they are understood.
© WASchaffer
33
Draft 5/11/2010
Techniques for Analysis
Note A
NOTE A. TECHNIQUES FOR DATA
ANALYSIS
Introduction
approximation to a self-sufficient
economy.
Now, with a few explanatory models
under our belts, let us take a look at a
couple of techniques for examining data.
The term “analysis” comes from the
Greek term for “a breaking up.” I like to
think of it as a “laying out” of essential
features for the better understanding of
some phenomenon or thing. The process
of analysis can involve a simple reordering of elements or it can involve a
complex statistical tool or maybe even a
model.
As pointed out in our discussion of
economic-base models, it is easily
converted for use in estimating export
employment. We simply assume that a
location quotient greater than one means
that an industry produces more than
expected in a self-sufficient economy
and thus would be an export industry.
On this assumption and location
quotients for a local economy, we built
an estimate of the economic base
multiplier for an economy.
Sometimes, the simple approach is the
best and can yield insights with great
benefits at small costs. Here, I would
like to re-visit our old friend the location
quotient and consider a new technique,
shift-share analysis. Both of these tools
have their critics in academic circles, but
both can be used to suggest strengths
and weaknesses in an economy and both
can point toward actions for developing
a community.
But all of these computations for
detailed industries can lead to questions
as well as answers. The analyst could
proceed to ask why the local economy
varies from the benchmark economy.
What resources are present or absent?
What opportunities for import
substitution are unexploited? What is
the apparent comparative advantage
signaled by a large location quotient?
Could it be an impediment or an
enhancement for future growth?
Location quotients
Obviously, this means that location
quotients are suggestive in nature -- they
point toward further analysis, and they
force us to proceed next to understand
production processes and to explore
comparative advantage and ways to
change the local economy and promote
growth.
A location quotient is commonly
defined as the ratio
LQi = (ei/e)/(Ei/E),
a-1
where ei is area employment in industry
i, e is total employment in the area, Ei is
employment in the benchmark economy
in industry i, and E is total employment
in the benchmark economy. Normally,
the "benchmark" economy is taken to be
the nation as the closest available
© WASchaffer
The popularity of location quotients
has lead to a large number of variations.
As Avrom Bendavid-Val points out in
his excellent practitioner’s book, the list
of derived measures includes “...
37
Draft 5/11/2010
Techniques for Analysis
Note A
coefficient of localization, coefficient of
specialization, index of diversification,
coefficient of redistribution, coefficient
of geographic association, coefficient of
participation, index of occupational
discrimination, coefficient of deviation,
friction ratio, and more. All of these
amount to little more than imaginative
applications of the basic location
quotient technique, computing a ratio of
ratios, in response to particular analytic
needs.” (Bendavid-Val 1983)
definitional manipulation of data
containing no explanation of phenomena
(that is, it is simply a way to organize
data) -- the explanation of change has to
come from further investigation.
But its major fault is that its
interpretation relies heavily on the level
of aggregation of the data used.
Nevertheless, it is a great and
inexpensive way to start a review of the
industrial structure of an area.
Now, to see how this tool works, let
us try various ways to estimate regional
employment in year 2 for industry i
(R2i), given knowledge of growth in
national employment in industry i (N2i/
N1i ), growth in total national
employment (N2./N1.), and even
knowledge of actual employment in
industry i in the region (R2i/R1i). How
could we proceed?
Advantages and disadvantages of
using location quotients are discussed in
Chapter 3 along with sources of data.
Shift-share analysis
Sometimes, it may be helpful to
increase your general knowledge of
change in the area in which you are
conducting a regional impact analysis.
For this task, “shift-share analysis” may
be appropriate. It is the most common
technique for breaking economic change
in areas into components is called “shiftshare analysis.”1
There are four alternative estimating
techniques consistent with shift-share
analysis. First, we could assume that
employment in industry i is the same in
period 2 as it is in period 1:
Although originating in the 1940’s,
the technique was introduced to frequent
use in 1960 by a team of economists
undertaking a massive study of regions
and economic growth (Perloff et al.
1960). Since then, it has been extended,
used, criticized and revived numerous
times. As a projection technique, it has
been abandoned by all (including the
U.S. Department of Commerce, a
staunch supporter in the 1960’s) but the
most faithful. As an expository
technique, it has enjoyed continued life.
Economists have criticized it as merely a
R2i = R1i
Here, we simply have neither knowledge
nor hope for growth.
Second, we could assume that the
local employment in industry i grows at
the same rate as does the national
economy:
R2i = R1i*N2./N1.
a-3
This, of course, is also a naîve
assumption. In effect, we have assumed
that both the industry and the economy
have grown at the same rate, that local
elements have each retained a constant
share of national growth (hence the
term).
1 A clear statement of this technique is contained in an
essay by Charles F. Floyd included as an appendix to
(Schaffer 1976). My comments rely on Professor
Floyd’s work as well as appendix B in (Jackson et al.
1981)
© WASchaffer
a-2
Third, we could assume that the local
employment in industry i grows at the
38
Draft 5/11/2010
Techniques for Analysis
Note A
same rate as does the employment in
industry i in the national economy:
R2i = R1i*N2i/N1i
Total change for each industry can be
summarized as:
a-4
TCi = NGi + IMi + RSi
a-9
This is a little better. We can reasonably
assume that local industries are subject
to the same demand pressures felt by
their competitors at the national level.
These elements can then be summed to
yield:
Fourth, we could assume that the local
employment in industry i grows at the
same rate as actually occurs (if we
somehow could know this!):
Demonstration that all of these equations
fit together to form this identity is left as
an exercise.
R2i = R1i*R2i/ R1i
TC =
NGi = R1i*N2./N1 - R1i
a-6
a-10
An area profile might be defined as an
economic description which may lead to
action or insight. Profiles may have
many purposes. They may be intended
to inform potential investors, to attract
visitors, to inform citizens about the
structure of their community, etc. The
challenge for an analyst is to break free
of the exclamatory hype associated with
promotional literature. You should
organize data to proclaim advantages
and accomplishments, to identify
problems and opportunities, or to
suggest future policy, as appropriate.
The following words and phrases are
random points derived from a general
reading of published and unpublished
profiles; I suggest a scanning of
collections of brochures, of Federal
IMi = R1i*N2i/N1i - R1i*N2./N1.
a-7
The last change is called the regionalshift effect (RSi) and shows the
additional change due to the specific
characteristics of the region itself:
RSi = R1i*R2i/ R1i - R1i*N2i/N1i
© WASchaffer
 R1i )
Thoughts on writing an area profile
The other two changes represent shifts
away from the national trend. The first
is called the industry-mix effect (IMi)
and shows the additional change due to
the growth characteristics of industry i:
RSi = R1i(R2i/ R1i - N2i/N1i)
2i
i
We should note that the terminology
used here, while it seems the most
common, is not unique. The technique
has been used and reinterpreted
frequently, leading to renaming of
effects. “Effects” are sometimes
“components.” The “industry-mix
effect” has been the “structural
component,” the “proportional shift,”
and the “industry-composition effect;”
and the “regional-share effect” has been
called the “differential shift” and the
“competitive effect.” (Floyd 1976)
a-5
The change in employment due to
national growth alone, assuming the
region gets its share, we call the national
growth effect (NGi):
IMi = R1i(N2i/N1i - N2./N1.)
i
i
Now, what if we try subtraction to
make equations a-3 through a-5 into
expressions of net change so that they
can be added? We can do this by
subtracting the right side of the
preceding equation from each and
naming the results.
NGi = R1i(N2./N1 - 1)
 TC  (R
a-8
39
Draft 5/11/2010
Techniques for Analysis
Note A
Reserve Economic Reviews, of the
World Wide Web, etc. for further
guidance:
What is the reason for the region’s
existence? Why do people live there?
What will happen in the next two,
three years -- expected new
industries or activities, departures,
significant events?
What is the region? Describe it
absolutely and relative to other
places.
How is the region defined? If it is a
political region, what is its economic
component. If it is a statistical region,
are there any unusual features?
Development policies stated by
authorities and in practice.
Problems
Trends in various indicators and
performance relative to nation or
other regions.
Geography and significant landscape
features, advantages, and limitations.
Demography: population, age and
ethnic composition if significant,
education, etc.
Why should the region be optimistic
or pessimistic about its future?
(Remember that you are an objective
analyst, not a promoter.)
Employment -- Have recent changes
been due to dramatic changes for
particular industries, or have they
occurred across the board?
Elements to include in a location
quotient analysis
Governments -- fragmented,
consolidated, etc.
A location-quotient analysis should be
carefully planned to take advantage of
modern spreadsheets such as Microsoft’s
Excel. The challenge is to lay out the
system so that you can sort and resort on
the basis of the values of location
quotients and on calculated surplus
employment and back into industry
order as needed.
Subregions -- definitions and
boundaries, uneven growth or
population distribution?
Economic base, before, now,
expected. Location quotient analysis.
How similar to the nation or state?
Should it change? Where is the
market for its exports? Where should
major imports originate? Are these
other markets stable or volatile (that
is, is the region subject to an
interregional business cycle)?
You should pay careful attention to
presentation of data in tabular format:
Does the table title clearly identify the
region under study, the year, and the
data?
Will the economic base help or hinder
in the future?
Are the columns clearly identified in
the caption?
What are the prospects for improving
the industry mix to get higher
personal incomes, less pollution,
whatever?
Does the stub identify all rows and
are sections specified and set apart?
Are columns wide enough for data?
Are decimals set consistently and at
the appropriate level for discussion?
Which industries lead the region’s
performance? Do they contribute to
cycles or to stability?
© WASchaffer
40
Draft 5/11/2010
Techniques for Analysis
Note A
In other words, does the presentation
meet the standards of good
scholarship and sound workmanship?
County Business Patterns) to see if
they yield similar results?
Did you use both quotients and
estimated excess employment?
Have you presented only data that is
relevant, excluding items that are
inconsequential?
Did you set up the tables and your
statements such that the reader could
easily check your results and even
make their own conclusions?
The written part of the analysis is also
important. Here are some points to
consider:
Did you speculate on apparently
questionable results?
At a very basic level, do you define
“location quotients” and how they are
used?
Did you explore changes over time?
(Since REIS data is since 1969 and
the CBP data is available in yearly
chunks on CD-ROM since 1986, we
might expect an alert and energetic
analyst to look at variation over time
either to show trends or to identify
data problems.)
Do you show awareness of your data
source, the level at which you work,
and problems which may be
associated with your data?
Did you experiment with several
sources (e.g. both the Regional
Economic Information System and
© WASchaffer
41
Draft 5/11/2010